Image

Mapping native forest using deep learning

2 May 2021

In this article, we introduce a deep learning pipeline that extracts information from high-resolution aerial data. The solution extends our current machine learning processes that use an object-based approach for identifying tree seedlings using data captured by a cessna-mounted imaging system.

In this example, we take implementation one step further and use a deep learning segmentation model that extracts information from high-resolution aerial data. We test the solution over an area that includes recently planted pines and pockets of established and regenerating native bush. The scaleable workflow developed uses data collected from the aforementioned airborne camera system and the results are a collaboration between Indufor and Scion’s Geomatics team.

Imagery used covers an area that was captured two-three months after planting at resolutions between 2.5 to 5 cm. The imagery collected was ingested into Indufor’s iterative deep learning framework designed to rapidly build high-quality datasets for deep learning.

A masked R-CNN neural network model pre-trained on the COCO dataset was trained on select areas of native forest and sprayed spots. The model builds on our pre-existing object detection model, used to identify spot spraying. Object detection models locate and represent objects with a bounding box, which suits features with relatively homogeneous size and shape. However, such a relatively crude method affects the model’s versatility, particularly when the targets are irregular and there is substantial variability between the geometry of targeted features (forests).

Comparatively, the two-class model combines the object detection algorithm with a semantic segmentation component which classifies each pixel in a bounding box. This enables sprayed spots and native patches to be accurately delineated (see the image below). The output can then be exported into common spatial formats for further analytics in a GIS and remote sensing environment. The detected sprayed spots are useful for forest managers to understand the quality of establishment. Automated mapping of the native vegetation is the first step in understanding and improving the biodiversity value of afforested sites. This is especially important as New Zealand looks to scale afforestation efforts to meet climate change targets.

We tested our model over a subset area of 16 ha. Within this area, we identified 13,295 discrete release spraying spots and 1,541 discrete blocks of native forest, both with an average confidence score of 99%. The confidence score dictates the confidence of the model in terms of the predicted feature. From these results, we also determined spot spraying occupied some 2.1 ha and native forest, 1.4 ha.

Click here to read the full article

Source and image credit: Indufor

Read More
Image

Kea Aerospace achieves milestone in autonomous flight

Kea Aerospace has successfully completed a landmark BVLOS (Beyond Visual Line of Sight) flight, demonstrating the capabilities of their advanced autonomous flight technology. The milestone UAS (Unmanned Aircraft System) flight, conducted within Kea Aerospace’s Special Use Airspace, marks significant growth ...
Read more
Image

€100 m investment in wood-based EV batteries

Stora Enso Oyj is preparing to invest about €100 million in a pioneering project to use wood in electric vehicle batteries and offer an alternative to components currently made in China, according to its chief executive officer. The funds would ...
Read more
Image

First long-range electric trucks to New Zealand

Windrose Technology will work with Auckland based Etrucks® to bring the first long-range electric trucks to New Zealand. The Windrose E1400 6×4 semi tractor features extreme aerodynamic efficiency, with an industry low coefficient of drag of 0.2755. Combined with the ...
Read more
Image

DC Equipment Canada return Madill to B.C.

DC Equipment has announced the opening of its new manufacturing facility in Prince George, marking the return of Madill equipment production to British Columbia. The first machines to be produced at this facility will be the Madill 3000B Log Loaders, ...
Read more
Image

QLD budget weak on future timber material needs for housing

Queensland’s peak body for the forest and timber industry is warning of a looming building materials crisis by a lack of policy and budget measures for increasing future timber supply to help meet the state’s rising housing demand. “While we ...
Read more
Image

Kenworth unveils futuristic SuperTruck 2

Kenworth showcased its SuperTruck 2 demonstrator vehicle at the Advanced Clean Transportation (ACT) Expo in Las Vegas last month, a project developed over six years in collaboration with the US Department of Energy’s (DOE) SuperTruck programme. The aim of the project ...
Read more
Image

Key drivers to decarbonise heavy transport fleets

The industry was buzzing once again for the Wood Transport & Logistics 2024 event that ran last week in Rotorua, New Zealand. Well over 200 forestry and log transport companies from across Australasia, America and Europe, came to Rotorua to ...
Read more
Image

Rocket Lab launches climate science satellites

Rocket Lab, a global leader in launch services and space systems, today launched the first of two back-to-back launches for NASA’s PREFIRE (Polar Radiant Energy in the Far-InfraRed Experiment) climate change-focused mission. ‘Ready, Aim, PREFIRE’ lifted-off from Rocket Lab Launch ...
Read more
Image

Hiringa launches hydrogen refuelling network

Taranaki-based green hydrogen developer Hiringa Energy has marked the opening of New Zealand’s first zero-emission green hydrogen refuelling network set to service heavy freight across the North Island. The initiative is the first of its kind in Australasia and was officially ...
Read more